Bicycle Computer Calibration

I'm an engineer, so you'll have to forgive me.  I recently had to change the batteries on my Cateye bicycle computer, so I decided to really do a better job in calibration.  Using the instruction number was giving me errors of approximately 10%, not even good for government work.

Using the Nike signs on the Lakefront path I was able to get within 0.01 mile on a 7 mile stretch.  I'm happy.

But here's the dilemma.  Are the signs that post the mileage accurate?  Going south from Du Sable Harbor to the Cultural Museum seems to be ok, as is the far northern section of the path.  The part between Navy Pier and Oak Street appears to be way off.

Anyone else have the same observation?

Views: 273

▶ Reply to This

Replies to This Discussion

Permalink Reply by Thunder Snow on June 14, 2013 at 7:18am

Rather than trusting signs, with no way of checking their accuracy, why not use the Chicago street grid for your measure?  I would think if you rode from the southeast corner of State & Madison up to the southeast corner of Chicago & State (or any other 8-block stretch that doesn't include Madison south to Roosevelt, which are older, shorter blocks) you'd have a very accurate mile to work with.

Permalink Reply by Michael J Blane on June 14, 2013 at 8:02am

Most of the bike computers have a formula in the manual for adjusting to different tire sizes. I've done my own in the past.

Mark the sidewalk (tire stem down) roll it at least one revolution, do the math; pretty easy. I usually did two revolutions (adjusting the formula) to help reduce the effects of any errors in my measuring.

Good Luck.

Permalink Reply by Doug Haynes on June 14, 2013 at 8:08am

This is the only way to get it dead on.

Mile markers are very rarely accurate to the degree you are looking for even on highways.

Michael J Blane said:

Most of the bike computers have a formula in the manual for adjusting to different tire sizes. I've done my own in the past.

Mark the sidewalk (tire stem down) roll it at least one revolution, do the math; pretty easy. I usually did two revolutions (adjusting the formula) to help reduce the effects of any errors in my measuring.

Good Luck.

Permalink Reply by Jeff Markus on June 14, 2013 at 8:47am

And to get even more confused remember you are counting revolutions and when you fill your bike pressure to the max suggested pressure the tire will roll on a larger diameter than when you are even 10% under inflated. So now that yer ready to just chuck it here is another alternative that will be 'accurate' regardless of bike, tire size or inflation but gives linear travel not actual bike related distances.

I spaek of a handheld GPS (global positioning system) unit.

Now it gets even more indistinct....not all units are created equal. I first bought a Garmin e-Trex unit to geocache with and was blessed by having picked up one of the 'faster' cheap handheld units, which means it 'refreshes' frequently updating position and data. But bcause the caching needed to be inserted by buttons it can be tedious so I bought a Magellan unit that loads from a USB port on my 'puter. Despite being a few bucks more or because it has more features to kick up costs the Magellan eXploist GC is noticably slower on the uptake. The speed of refresh affects the ability and accuracy of distance measurements since the positioning intervals determine the measured course.

So the Garmin when strapped to me handle bars or in my pocket gives me more accurate speed, distance and location at shorter intervals (and has a more easily read B&W display) that I can display in large enuf digits to read while in motion. This unit has been handy for evaluating commuting distance, speeds and the like while I broke in my e-bike.

So if ya need accurate point to point distance measurements go GPS or if its just maintenance intervals for your bike then get asclose as ya can with the understanding that many variables affect bike computers since they don't read directly to the ground just to the turns of the wheel which is affected by tire/wheel size, inflation, even loaded weight (how much did you have for lunch?).

Permalink Reply by David P. on June 14, 2013 at 8:49am

Based on the mindset implied by his background (I'm an engineer too) I'm going to assume that Jeff did calibrate it this way - that in fact it was calibrating using direct rolling-circumference measurements rather than the calibration table that came with the computer that led to the better results he sees. I understand him to really only be asking, "can you really trust the mileage markers on the LFP?" To this, I would say "no" and agree with the suggestion to use a portion of the grid near downtown (it can be irregular in places) to check the results.

Permalink Reply by Anne Alt on June 14, 2013 at 8:55am

I've done that and compared it against a known 1 mile segment of road (as measured by car odometer and the Chicago street grid).

notoriousDUG said:

This is the only way to get it dead on.

Mile markers are very rarely accurate to the degree you are looking for even on highways.

Michael J Blane said:

Most of the bike computers have a formula in the manual for adjusting to different tire sizes. I've done my own in the past.

Mark the sidewalk (tire stem down) roll it at least one revolution, do the math; pretty easy. I usually did two revolutions (adjusting the formula) to help reduce the effects of any errors in my measuring.

Good Luck.

Permalink Reply by ilter on June 14, 2013 at 4:11pm

"The part between Navy Pier and Oak Street appears to be way off."

Since you say way off.. did you go on the sidewalk behind the (?)Lakefront Condo, or stay on the path & go around that condo and near the Navy Pier entrance?  Even if you stayed on the path, the mile markers might easily be off with all the street and sidewalk crossings near Navy Pier.

Permalink Reply by Michael A on June 14, 2013 at 9:55pm

try calibrating it next to a gps app from a cell phone or garmin

RSS

Groups

© 2008-2016   The Chainlink Community, L.L.C.   Powered by

Disclaimer  |  Report an Issue  |  Terms of Service